KR20020019932A - Multi-rate digital signal processor for vibrating conduit sensor signals - Google Patents

Abstract

A digital signal processor 291 that determines the characteristics of the material flowing through the conduits 103A-103B. The digital signal processor 201 of the present invention receives signals at a first sample rate from two pick-off sensors 105-105 'installed at two different points along the flow. This signal is converted into a digital signal. The digital signal is decimated to the desired sample ratio at the first sample rate. The frequency of the received signal is then determined from the digital signal of the desired sample ratio.

Description

[0001] MULTI-RATE DIGITAL SIGNAL PROCESSOR FOR VIBRATING CONDUIT SENSOR SIGNALS [0002]

It is known to use a Coriolis effect mass flow meter to measure the flow rate and various information about the material flowing through the conduit in the flow meter. second. this. A typical Coriolis flow meter is disclosed in US Pat. Nos. 4,109,524, issued May 8, 1978 to Smith et al., U.S. Patent No. 4,491,025, and U.S. Patent No. 31,450, issued Feb. 2, 1982. These flow meters have one or more straight or curved conduits. The shape of each conduit in the Coriolis mass flow meter has a set of natural vibration modes, which may be simple bending, twisting, or a combination thereof. Each conduit is driven to oscillate in resonance in any of these eigenmodes. The material flowing into the flow meter from the pipeline connected to the inlet side of the flow meter is delivered through the conduit (s) and exits the flow meter through the outlet side of the flow meter. The natural vibration mode of a vibrating, material-filled system is determined in part by the combined mass of the material flowing in the conduit and the conduit.

When there is no material flowing through the flow meter, all points along the conduit will either move in the same phase by the applied driving force or flow with a small initial compensating phase offset that is correctable. As the material begins to flow, the Coriolis force causes each point in the conduit to have a different phase. The phase at the inlet side of the conduit is delayed relative to the driver, while the phase at the outlet side of the conduit leads the driver. A pick-off sensor is installed in the conduit to produce a sine signal indicative of the motion of the conduit. The signal output from the pick-off sensor is processed to determine the phase difference between the pick-off sensors. The phase difference between the two pick-off signals is proportional to the mass flow rate of the material flowing through the conduit.

The Coriolis flow meter has a transmitter that generates a drive signal to drive the actuator and determines the mass flow rate and various characteristics of the material from the signal received from the pick-off sensor. Conventional transmitters are made of analog circuitry designed to generate drive signals and detect signals from pick-off sensors. Analog transmitters have been optimized over the years and manufacturing costs are relatively low. Therefore, it is desirable to design a Coriolis flowmeter capable of using a conventional transmitter.

Conventional transmitters have a problem that they have to work on signals with a narrow range of operating frequencies. The range of such operating frequencies is typically 20 Hz to 200 Hz. This limits designers to such a narrow range of drive frequencies. Moreover, this narrow range of operating frequencies makes it impossible to use conventional transmitters for some flow meters, such as a linear tube Coriolis flowmeter that operates at a high range of frequencies, such as, for example, 300-800 Hz. Linear tube flow meters operating at 300-800 Hz tend to be less sensitive to the Coriolis effect used to measure mass flow rates. Therefore, it is necessary to measure the phase difference between the sensors more precisely in order to calculate the mass flow rate.

In order to use one type of transmitter for several different designs of Coriolis flowmeters operating at several different frequencies, manufacturers of Coriolis flowmeters use a digital signal processor to generate a drive signal and process the signal received from the pick-off sensor Lt; / RTI > A higher requirement for the measurement resolution and accuracy imparted to the analog electronic elements by the flow meter operating at high frequencies, such as a straight tube design, is eliminated by digitizing the signal from the pick-off when the signal is received by the transmitter A digital signal processor is preferred. In addition, instructions for signaling processing using a digital processor can be modified to operate at several different frequencies.

However, digital signal processors have several disadvantages over conventional analog circuit transmitters. The first problem with digital signal processors is that the circuit elements of the digital processor are more complex and therefore costly to manufacture. Second, digital signal processors require a circuit board with a large surface area, which is a problem when there is insufficient space in the flowmeter design. Third, digital signal processors require more driving power than analog circuits. Power consumption is particularly problematic when the processor must operate at the maximum clock rate to provide all the calculations needed to process the signal and generate mass property measurements such as mass flow rate. For all these reasons, there has been a need for a digital signal processor that can be applied to several flow meter designs, is low in manufacturing cost, and can reduce the amount of power required to perform the required calculations.

The present invention relates to a signal processor for an apparatus for measuring a characteristic of a material flowing through at least one conduit vibrating in the apparatus. More particularly, the present invention relates to a digital signal processor that performs calculations to determine the frequency of a received signal from a pick-off sensor that measures the oscillation frequency of the conduit.

The invention can be understood from the following detailed description and the accompanying drawings.

1 shows a Coriolis flowmeter having a digital transmitter for performing a multiple bipy-off signal processing process of the present invention.

Figure 2 shows a block diagram of a digital signal transmitter.

3 shows a flow chart of the operation performed by the digital transmitter.

Figure 4 shows a flow diagram of a process for generating data from a signal received from a pick-off sensor.

5 shows a process for performing decimation of a signal sample from a pick-off.

6 shows a process for calculating an estimated frequency of a signal received from a pick-off.

FIG. 7 illustrates a process for performing a high-to-low frequency selection on a received signal.

8 shows a process of demodulating a received signal.

Figure 9 illustrates a method for determining data on flow tube vibration from a received signal.

These and other problems are overcome by the provision of a multi-rate digital signal processor of the present invention and progress in this field. The present invention consists of a processing process that is stored in a memory and executed by a processor to process a signal received from a pick-off on a vibration conduit. The process steps of the present invention provide many advantages that make it possible to use a single type of digital signal processor in various types of Coriolis flow meters.

A first advantage of the present invention is that the processing steps do not lose accuracy despite using fixed operations instead of floating point operations. A second advantage of the processing process of the present invention is that processing processes can be performed on a variety of low-cost, low-power digital signal processors such as Texas Instruments TM3205xx, Analog Devices ADSP21xx, or Motorola 5306x. The instructions of the processing process of the present invention are small enough to remain in the internal memory of the digital signal processor, eliminating the need for a fast access external memory that increases cost, power consumption and circuit board space for the transmitter. These processes have a small number of computational configurations, thus improving the portability of the processing process between low cost processors.

A third advantage is that the computational requirements of the treatment process are minimized. The reduction of the computational requirements causes the digital signal processor to operate at a lower clock rate than the processor ' s maximum clock rate, thereby reducing power consumption of the processor.

The transmitter performing the processing process of the present invention has the following electronic elements. An analog signal from the pick-off attached to the sensor is transmitted to an analog / digital (" A / D ") converter. The converted digital signal is applied to a standard digital processor. A digital processor is a processing unit that implements machine readable instructions stored in memory coupled to the processor via a bus. A typical digital processor has a fixed storage device (ROM) storing instructions for performing the required processing steps, such as the processing steps of the present invention. The processor is also coupled to a random access memory (RAM) that stores instructions for the currently executing process and data necessary to perform the process. The processor may also generate a drive signal for the Coriolis flowmeter. To apply a drive signal to the drive system, a digital processor may be coupled to a digital / analog (D / A) converter that receives the digital signal from the processor and applies an analog signal to the drive system.

The process of the present invention performs the following functions to determine the frequency of the signal received from the pick-off sensor and the phase difference between the signals. First, signals are received from the pick-off sensor at a first sample rate. Off means the amount of input received from the pick-off that is used to characterize the signal from the sample contrast peak. The signal is then decimated to the desired sample ratio at the first sample rate. Decimation simply means converting from the first sample number to a smaller number of samples.

Decimation is performed to increase the resolution of the sampled signal to provide a more precisely calculated signal frequency for each signal. The frequency of each signal is then determined.

The following steps are additionally carried out so that the same treatment process can be used for other flow meters having different frequencies. An estimate of the oscillating frequency of the flow meter is calculated. The estimated frequency is then used to demodulate the signals from each pick-off into an I component and a Q component. The I and Q components are then used to translate the signal to a center frequency when the operating frequency of the signal is greater than the transition frequency. After demodulation, the signal can be decimated again to improve the resolution of the signal again.

The dominant frequency of the signal is then isolated and measured precisely. The movement to the zero frequency is then calculated for the I component and the Q component of the signal. At this time, each component is decimated again to increase the measurement accuracy. The frequency band of each signal can then be narrowed as desired by appropriate low-pass filtering. A complex correlation is then performed to determine the phase difference between the signals.

According to one aspect of the present invention, the present invention provides a method for measuring a signal received from a first pick-off sensor and a second pick-off sensor for measuring vibration of a conduit, A signal processor is used to process in the following manner. A processor receives a sample of the signal at a first sample rate from the first pick-off sensor and the second pick-off sensor. The sample is decimated from the first sample ratio to a desired sample ratio. The vibration frequency for the conduit in the first pick-off sensor and the second pick-off sensor is determined from a sample of the signal at the desired sample ratio.

Another aspect of the present invention is to demodulate the signals from the first pick-off sensor and the second pick-off sensor to move to a center frequency.

Another aspect of the invention is to calculate the normalized frequency of the signal.

In another aspect of the present invention, the demodulation step is performed in the following manner. A normalized pulse of the normalized frequency of the signal is calculated. A scalar multiplication of the signal from the normalized pulsation and the first pick-off sensor and the second pick-off sensor is calculated to move the signal to the center frequency.

In another aspect of the present invention, the step of calculating the normalized frequency is performed in the following manner. The signal is demultiplexed into an I component and a Q component. The I component is integrated and the Q component is integrated. The I and Q components are multiplexed to produce a digitally integrated signal. A ratio between the amplitude of the signal and the amplitude of the digitally integrated signal is calculated to produce the normalized frequency of the signal.

Another aspect of the present invention is that the step of calculating the normalized frequency is performed by applying the integrated Q component to the compensator in response to the integration step prior to the multiplexing step.

Another aspect of the invention is that the step of calculating the normalized frequency is performed by applying the integrated I component to the compensator in response to the integration step prior to the multiplexing step.

In another aspect of the present invention, the step of determining the vibration frequency of the conduit is performed in the following manner. The normalized frequency of the signal is determined. The normalized frequency of the signal is modulated. A complex demodulation of the signal is performed using the modulated normalized frequency to determine the oscillation frequency.

In another aspect of the present invention, the step of determining the vibration frequency of the conduit is performed in the following manner. The demodulated signal is decimated. A complex correlation of the signals is performed to determine the phase difference between the signals.

Another aspect of the present invention is to determine a phase difference between a signal from the first pick-off sensor and a signal from the second pick-off sensor.

Another aspect of the invention is to determine the characteristics of the material flowing through the conduit in response to determining the frequency of the signal from the first pick-off sensor and the signal from the second pick-off sensor.

Another aspect of the invention is that one of the properties is the mass flow rate of material flowing through the conduit.

Another aspect of the invention is that one of the properties is density.

In another aspect of the present invention, the step of decimating the signal from the first pick-off sensor and the signal from the second pick-off sensor is performed in the following manner. A first decimation is performed with an intermediate sample ratio from the first sample ratio in response to receipt of a signal from the first and second pick-off sensors.

In another aspect of the present invention, the decimating step is performed in the following manner.

The signal is demodulated into an I component and a Q component. A second decimation of the I component and the Q component of the sample of the signal is performed at a final sample ratio in the intermediate sample ratio in response to demodulation of the signal.

Coriolis flowmeter overview - Figure 1

Figure 1 shows a Coriolis flowmeter 5 with a Coriolis flowmeter assembly 10 and a transmitter 20. The transmitter 20 is connected to the meter assembly 10 via a lead 100 to provide density, mass flow rate, volume flow rate, total mass flow rate and consolidation density via path 26. A specific Coriolis flowmeter structure has been described, although it will be clear to those skilled in the art that the present invention may be practiced in conjunction with any device having a vibrating conduit to measure the properties of the material. A second example of such a device is a vibration tube density meter that does not have the additional measuring capability provided by the Coriolis mass flow meter.

The flow meter assembly 10 includes a pair of flanges 101, 101 ', a manifold 102, and conduits 103A, 103B. A driver 104 and pick-off sensors 105 and 105 'are connected to the conduits 103A and 103B. Brace bars 106 and 106 'form the axes W and W', with each conduit pivoting about the axis.

When the flow meter 10 is inserted into a pipeline system (not shown) that transports the process material to be measured, the material enters the flow meter assembly 10 through the flange 101 and passes through the manifold 102, Flows into conduits 103A and 103B and flows through conduits 103A and 103B back into manifold 102 where they exit flowmeter assembly 10 through flange 101 '.

The conduits 103A and 103B are selected and properly mounted on the manifold 102 to have approximately the same mass distribution, moment of inertia and modulus of elasticity for each of the bending axes W-W and W'-W '. The conduits extend outwardly from the manifold in a generally parallel manner.

The conduits 103A-103B are driven by the driver 104 in opposite directions with respect to the respective bending axis W, W ', which is referred to as the first out of phase bending mode of the flow meter. Actuator 104 may be constructed of well known devices such as a magnet mounted on conduit 103A and an opposing coil mounted on conduit 103B and an alternating current is passed through the device to vibrate both conduits. A suitable drive signal is applied to the driver 104 by the flow meter electronics element 20 via the lead 110. [

The transmitter 20 receives the left and right velocity signals respectively appearing on the leads 111 and 111 '. The transmitter 20 generates a drive signal appearing on the lead 110 to cause the driver 104 to vibrate the tubes 103A and 103B. The transmitter 20 processes the left and right velocity signals to calculate the density and mass flow rate of the material through the meter assembly 10. This information is applied to path 26.

It is well known to those skilled in the art that the Coriolis flow meter (5) is very similar in structure to the vibrating tube density meter. Also, the vibrating tube densitometers use a vibrating tube in which a fluid flows or, in the case of a sample-type densitometer, a fluid is accommodated. The vibrating tube densitometers also employ a drive system that provides excitation for vibrating the conduit. Since density measurement only requires frequency measurement and phase measurement is not required, the vibrating tube densitometer typically uses only a single feedback signal. The present invention described herein applies equally to oscillating tube densitometers.

Digital Transmitter (20) - Figure 2

Figure 2 shows the components of the digital transmitter 20. The paths 111 and 111 'transmit the left and right velocity signals from the meter assembly 10 to the transmitter 20. The velocity signals are transmitted to the analog / digital (A / D) converter 203 of the flow meter electronic element 20. The A / D converter 203 converts the left and right speed signals into digital signals usable in the processing device 201 and transmits the digital signals through the paths 210-210 '. Although shown as a separate element, the A / D converter 203 may be a single converter that provides two converters so that signals from both pick-offs can be simultaneously converted, such as an AK4516 16-bit codec chip. This digital signal is passed to the processor 201 by a path 210-201 '. It will be appreciated by those skilled in the art that any number of other sensors, such as RTD sensors for determining the temperature of the pick-off and flow tube, may be coupled to the processor 201.

The drive signals are transmitted through path 212, which applies a signal to a digital to analog (D / A) converter 202. The D / A converter 202 is a common D / A converter, which may be a separate D / A converter or integrated into a stereo CODEC chip such as the standard AKM4516. Another common D / A converter 202 is the AD7943 chip. The analog signal from the D / A converter is sent to the driving circuit 290 via the path 291. Then, the drive circuit 291 applies a drive signal to the driver 104 through the path 110. [ The path 26 carries a signal to an input / output means (not shown) which allows the transmitter 20 to receive data from or transmit data to the driver.

The processing unit 201 is a group of processors that reads instructions from a micro-processor, processor, or memory and executes instructions to perform various functions of the flow meter. In a preferred embodiment, the processor 201 is an ADSP-2185L microprocessor manufactured by Analog Devices, Inc. The functions performed include, but are not limited to, calculating the density of materials that can be stored in the fixed memory (ROM) 220 as instructions, calculating the mass flow rate of the material, and calculating the volumetric flow rate of the material. This information and instructions for performing various functions are stored in random access memory (RAM) The processor 201 reads or writes the work in the RAM storage device 230 via the path 231. [

Outline of the process performed by the digital transmitter 20 - Fig. 3

Figure 3 is an overview of the functions performed by the digital transmitter 20 to operate the Coriolis flowmeter 5. The process 300 begins by the transmitter 20 generating a drive signal at step 301. A drive signal is then applied to the driver 104 via path 110. In step 302, the digital transmitter 20 receives a signal from the pick-off 105, 105 'responsive to the vibrations of the flow tubes 103A-B as the material flows through the flow tubes 103A-B . In step 303, data relating to the signal such as the frequency of the signal and the phase difference between the signals is performed by the digital transmitter 20. Next, in step 304, information on the properties of the material flowing through the flow tubes 103A and 103B from the data, for example, mass flow rate, density, and volume flow rate, is calculated. Process 300 is then repeated as long as Coriolis flow meter 5 operates in the pipeline.

Data generation process for the pick-off signal according to the present invention - Fig. 4

4 shows a process 400 for generating data, such as the signal frequency, for a signal received from a pick-off 105, 105 'measuring the oscillation of the flow tubes 103A-B in the Coriolis flowmeter 5 have. The process 400 provides several advantages that are advantageous for use in the digital transmitter 20. The first advantage of the process 400 is that there is no degradation in accuracy despite the use of fixed-point arithmetic instead of floating-point arithmetic. This allows the use of low-cost, low-power processors such as the TMS3205xxx manufactured by Texas Instruments, the ADSP21xx manufactured by analog devices, or the 5306x manufactured by Motorola. A second advantage is that the memory need for instructions for process 400 is small enough to remain within the processor's internal memory, thus eliminating the need for a high-speed bus between the processor and external memory. Allowing the processor to operate substantially below the maximum clock ratio because the need for computation is reduced by the processor 400. [

The process 400 begins by decimating the sample rate of the signal received from the pick-off at step 401 to a second lower sample rate from the first sample. In a preferred embodiment, the signals are decimated from a first sample rate of 48 kHz to a second sample rate of 4 kHz. Decimation of the sample ratio increases the resolution of the signal, which increases the accuracy of the calculations described in FIG. 5 below. In a preferred embodiment, the sample ratio reduction from 48 kHz to 4 kHz increases the resolution of the sample from B bits to B + 1.79 bits.

In step 402, an estimate of the signal frequency is calculated from the sample signal. A preferred process for calculating the estimated signal frequency is shown in Fig. Then, in step 403, the estimated signal frequency is used to demodulate the received signal. The process of demodulating the digital signal is shown in Figs. In step 404, a second decimation of the sampled signal is performed. This second decimation reduces the signal sample to a third sample ratio at the second sample rate to increase the resolution of the sampled signal. In this preferred embodiment, the reduction is achieved from a sample rate of 4 kHz to a sample rate of 800 Hz, which increases the resolution to B + 2.95 bits. This reduction is performed in the same manner as in step 401. [

After performing the second decimation in step 404, a calculation is made based on the received signal. This is performed as part of the process shown in FIG. After the noise is removed, in step 405 the frequency of the signal from each pick-off is determined. In step 406, the phase difference between the signal from the first pick-off and the signal from the second pick-off is determined. Then, in step 407, the amplitude of each signal is determined. The process 400 is then repeated until the flow meter is in operation or the process 400 is terminated.

Decimating process of sample rate of signal from pick-off - Figure 5

Figure 5 shows a process for decimating the proportion of samples received from a pick-off. The same process is used for each of the decimations performed in steps 401 and 404 and in the process of determining the frequency of the signal. In each of these steps, the process 500 is performed separately for the signal from each pick-off. The difference between the decimations performed at each step is the length of the input data vector, as described below.

The decimation described in process 500 is performed using block processing, where the size of the input vector is equal to the decimation ratio. The decimation ratio means the amount by which the sample frequency is reduced by decimation. Block processing has operational advantages because it only has to be repeated at the output data rate, not at the input data rate. The basic principle of iterative block filtering is that state variables,

x _{k + 1} = A * x _k + B * u _k ;

y _k = C * x _k + D * u _k ;

here:

A, B, C, D = matrix representing the state of the system;

x _k = N + 1 state vector at time k;

u _k = input; And

y _k = Output representing the decimated signal.

Through induction, you can see that:

When decimating a signal with a factor M, only each Mth sample will be retained. Therefore, the last output row of the matrix can be eliminated by the following equation.

From the above it can be seen that the number of accumulation / multiplication operations for a single iteration of the above equation is:

N _MAC = (N + 1) * (N + M)

here:

N _MAC = number of accumulation / multiplication operations

N = degree of matrix A; And

M = Block size equal to the decimation ratio of the process.

Thus, the computational burden of performing decimation,

R _SVD = F _out * N _MAC

R _SVD = calculated load of processor; And

F _out = represents the filter output rate.

The memory required to perform the decimation is as follows:

A memory (ROM) for storing each filter coefficient that can be read only;

A memory (RAM) for storing a filter state vector (x _k ) to be read-write; And

Input block buffer memory (read-write).

5 shows a decimation process using the block processing method.

Process 500 begins by accepting m samples as a buffer in step 501 to generate an input block. The input block is then multiplied in step 502 with a state vector. Then, in step 503, a result of outputting each m-th sample to be used for another calculation is output. The process 500 ends after step 503.

Process of estimating frequency of received signal - Fig. 6

Process 600 is a process for estimating the frequency of a received signal to demodulate a signal in a subsequent process step. Process 600 must be completed at step 403 before the signal can be demodulated. Subsequent demodulation is described below and shown in FIG.

A process 600 for estimating the frequency of a signal is shown in FIG. This process 600 is performed for any one of the received signals. The process 600 begins by demultiplexing the sampled digital signal into an in-phase (I) and a quadrature (Q) component in step 601. Then, in step 602, digital integration is performed on the I component and the Q component of the signal. In step 603, a signal compensation is calculated for the integrated signal. Then, in step 604, the signal components are multiplexed to produce a digitally integrated signal. Then, in step 605, the ratio between the original signal and the digitally integrated signal is calculated. The ratio provides an estimate of the signal frequency that can be used to demodulate the signal in process 700.

The process 600 uses a fixed coefficient filter to estimate the frequency. Therefore, an iterative algorithm is not required. Since the iteration is not used, the process 600 always converges. Process 600 also rapidly tracks changes in frequency. The estimated frequency at the end of process 600 is given by:

here,

F _est = estimated frequency;

ω _est = normalized pulsation; And

F _s = frequency of the sample.

Process for a high-low frequency selector - Fig. 8

Process 800 of FIG. 8 illustrates an optional process that may be performed between the frequency estimate and demodulation of the signal. A high-to-low frequency selector is needed to determine the frequency of interest. The process 1000 for accurately measuring the signal frequency (see FIG. 9) shows that it converges more slowly within the estimated bias and normalized frequency ranges:

here:

F ₀ = normalized frequency of the signal.

From this equation it can be clearly seen that the process of determining the frequency is not accurate if the sample rate is 4 kHz and the frequency of the signal being measured is as low as 20 Hz. The process 800 heals this situation so that the process 1000 is used over a broad frequency band. This is accomplished by determining in a way that the process will operate at high or low frequencies. The process 800 begins at step 801 by determining if the estimated frequency is less than or equal to the reference frequency. In a preferred embodiment, the reference frequency is selected to be 250 Hz. 250 Hz is chosen because it is the frequency between the conventional flow tube and the normal operating frequency of the straight flow tube.

If the actual estimated frequency is lower than the reference frequency, the estimated frequency is returned to zero. If the actual estimated frequency is greater than the reference frequency, the estimated frequency is calculated as the actual estimated frequency minus 120 Hz.

Demodulation Process of Received Signal-Figure 7

7 shows a process 700 for demodulating a received digital signal. This process uses the estimated frequency calculated in process 600 (FIG. 6) or in process 800 (FIG. 8). Process 700 begins by calculating normalized pulsations, represented by the following equation, at step 701: < RTI ID = 0.0 >

here:

ω _d = normalized pulsation;

F _d = estimated frequency; And

F _s = frequency of the sample.

The 'twiddle' coefficient, which is a real number in step 701, is calculated according to the following formula:

W _k = cos (? _D k)

At this time, x 硫 = A 硫 cos (ω ₀ k + φβ)

here,

? = signal received from any one of the pick-off sensors.

In step 702, a 'twiddle factor' and a scalar product of the actual received signal are calculated by the following formula:

.

Note that if process 800 yields a low mode low frequency mode with an estimated frequency of zero, y? (K) = x? (K). Otherwise, the modulated output signal has the following two components:

However, this can be healed by the double decimation described below. The first component corresponding to the negative in the above equation is an important signal. The second signal corresponding to the plus sign in the above equation will be filtered out in the next decimation process in step 905 of process 900 (FIG. 9).

Process for generating data from a received signal - Fig. 9

9 shows a process 900 for generating data relating to a signal received from a sensor. The process 900 begins by calculating the application of a notch filter parameter calculated in the following manner at step 901: It is known that the notch filter variable is a single adaptive variable that can be expressed as:

Where a <1 is a convergence variable that adjusts the bandwidth of the filter and a ₁ is a variable that is searched through adaptation. Assume a ₁ <2 and note that:

a ₁ = -2 cos (ω)

z? = exp (? j?)

here:

z = zero point;

j = constant; And

ω = signal.

Thus, the polarity of the signal is expressed by the following formula:

p? =? exp (? j?)

In step 902, a ₁ is calculated for each signal. a ₁ is computed using one of many conventional algorithms such as RLS, RML, or SGN. This minimizes signal energy.

In step 902, the signal frequency of each signal is determined. To determine the frequency of the signal, the normalized frequency for the decimated signal is determined using the following formula:

here:

F ₀ = normalized frequency for the decimated signal; And

a ₁ is the current adaptation value of the notch filter variable.

The frequency of the signal can then be determined by multiplying the decimation frequency by the normalized frequency in step 902 (F _t = F ₀ x F _s ). In step 903, quadrature demodulation is performed on the signal. The quadrature demodulation is performed by selecting the demodulated signal as follows: the dominant frequency of the signal is shifted to zero. In process 903, a scalar product of the demodulated signal and the received signal is calculated. The demodulated signal is expressed in the following manner:

here:

ω ₀ = pulsation of the modulation signal;

F ₀ = frequency of the normalized signal calculated in step (902).

The received signals, as described above, may be represented as:

x? (k) = Acos (? ₀ k +??); ? = each signal from the pick-off sensor 105-105 '.

From the above equations, the output of quadrature demodulation is:

In order to further improve the signal resolution, a decimation is performed in step 904. In a preferred embodiment, such a decimation is an a x 40 decimation performed on both the I and Q components of the received signal. This decimation simplifies the result of complex quadrature demodulation as follows:

After the decimation is performed, the phase difference of the signal is performed in step 905. [ In an exemplary embodiment, the phase difference is calculated as follows. From either the left or the right pick-off sensor, the first signal received is conjugated according to the following formula.

The signal is then multiplied with the second signal to perform a complex correlation between the pick-off signals, as shown in the following equation.

Thus, the phase difference is given by:

The phase difference can then be used to calculate the mass flow rate of the material flowing through the flow meter and many other properties.

The processing steps for determining the data relating to the signals received by the transmitter 20 and the digital transmitter 20 for the Coriolis flowmeter 5 have been described so far. It is contemplated that other digital signal processors and processing processes may be designed that literally or equivalently infringe the present invention set forth in the following claims.

Claims (31)

Off signals from the first pick-off sensor 105 and the second pick-off sensor 105 'that measure the vibrations of the conduits 103A-103B, 1. A method (400) for processing using a digital signal processor (201) to output information, the method (400) Receiving (302) a sample of the signal at a first sample rate from the first pick-off sensor (105) and the second pick-off sensor (105 '); Decimating the sample from the first sample ratio to a desired sample ratio (401); And Determining (405) a vibration frequency for the conduit at the first pick-off sensor (105) and the second pick-off sensor (105 ') from a sample of the signal at the desired sample ratio . The method of claim 1, wherein the method (400) Further comprising the step of demodulating (403) the signal from the first pick-off sensor (105) and the second pick-off sensor (105 ') to a center frequency. 3. The method of claim 2, wherein the method (400) And calculating (600) a normalized frequency of the signal. 4. The method of claim 3, Calculating (701) a normalized pulse of the normalized frequency of the signal; And And calculating (703) a scalar multiplication of the normalized pulsation and signals from the first pick-off sensor and the second pick-off sensor to move the signal to the center frequency. . 4. The method of claim 3, wherein calculating the normalized frequency comprises: Demultiplexing the signal into an I component and a Q component (601); Integrating (602) the I component; Integrating the Q component (602); Multiplexing (604) the I and Q components to produce a digitally integrated signal; And And computing (605) a ratio between the amplitude of the signal and the amplitude of the digitally integrated signal to produce the normalized frequency of the signal. 6. The method of claim 5, wherein calculating the normalized frequency comprises: Further comprising applying (603) the integrated Q component to the compensator in response to the integrating step (602) prior to the multiplexing step (604). 6. The method of claim 5, wherein calculating the normalized frequency comprises: Further comprising applying (603) the integrated I component to a compensator in response to the integrating step (602) prior to the multiplexing step (604). The method of claim 1, wherein determining the oscillation frequency of the conduit comprises: Determining (901) a normalized frequency of the signal; Modulating (902) the normalized frequency of the signal; And And performing (903) complex demodulation of the signal using the modulated normalized frequency to determine the vibration frequency. 9. The method of claim 8, wherein determining the oscillation frequency of the conduit comprises: Decimating the demodulated signal (904); And performing (905) a complex correlation of the signals to determine a phase difference between the signals. The method according to claim 1, And determining (910) a phase difference between a signal from the first pick-off sensor and a signal from the second pick-off sensor. The method according to claim 1, Determining a characteristic of a material flowing through the conduit in response to determining a frequency of a signal from the first pick-off sensor and a signal from the second pick-off sensor. 12. The method of claim 11, wherein one of the properties is a mass flow rate of material flowing through the conduit. 12. The method of claim 11, wherein one of the characteristics is density. The method of claim 1, wherein the step (401) of decimating the signal from the first pick-off sensor and the signal from the second pick- And performing (401) a first decimation with an intermediate sample ratio from the first sample ratio in response to receiving the signals from the first and second pick-off sensors. 15. The method of claim 14, wherein the decimating comprises: Demodulating (403) the signal into an I component and a Q component; And And performing (404) a second decimation of an I component and a Q component of a sample of the signal with a final sample ratio at the intermediate sample ratio in response to demodulation of the signal. Receive signals from sensors 105-105'attached to conduits 103A-103B that are vibrated by driver 104 and transmit the conduits 103A-103B from signals received from sensors 105-105 ' Wherein the signal indicates motion of the conduit from two or more points, and wherein the device (20) An analog-to-digital converter (203-203 ') for converting a sample of the signal into a digital sample of the signal; And And a processor 201, Wherein the processor (201) Decimating (401) the sample from the first sample ratio to a desired sample ratio, To determine (405) the vibration frequency for the conduit (103A-103B) from the first pick-off sensor (105) and the second pick-off sensor (105 ' Device (20). 17. The apparatus of claim 16, wherein the processor (201) is further configured to move signals from the first pick-off sensor (105) and the second pick-off sensor ). 18. The apparatus (20) of claim 17, wherein the processor (201) is configured to perform the movement of the signal by calculating (701) the normalized frequency of the signal. 19. The method of claim 18, wherein the processor (201) further calculates (701) normalized pulsation of the normalized frequency of the signal, and calculates the normalized pulsation and the first (20) configured to move the signal by calculating (703) a scalar product of the signal from the pick-off sensor (105) and the second pick-off sensor (105 '). The method of claim 18, wherein the processor (201) demultiplexes (601) the signal to an I component and a Q component, integrates (602) the I component, integrates (602) (604) the I and Q components to produce an integrated signal, and to multiply the ratio of the amplitude of the signal to the amplitude of the digitally integrated signal to produce the normalized frequency of the signal (20) configured to calculate the normalized frequency. 21. The apparatus of claim 20, wherein the processor (201) is configured to calculate the normalized frequency by applying the integrated Q component to a compensator in response to the integrating step (602) prior to the multiplexing step (604) 20). 21. The apparatus of claim 20, wherein the processor (201) is configured to calculate the normalized frequency by applying the integrated I component to a compensator in response to the integrating step (602) prior to the multiplexing step (603) 20). 17. The system of claim 16, wherein the processor (201) (901) determining a normalized frequency of the signal; Modulate (902) the normalized frequency of the signal; And (903) complex demodulation of the signal using the modulated normalized frequency to determine the frequency of oscillation, And to determine the vibration frequency of the conduit. 24. The system of claim 23, Decimating the demodulated signal (904); By performing (905) a complex correlation of the signals to determine the phase difference between the signals, And determine the oscillation frequency of the conduit. 17. The apparatus (20) of claim 16, wherein the processor (201) is further configured to determine (906) a phase difference between a signal from the first pick-off and a signal from the second pick-off. 17. The system of claim 16, (20) configured to determine a characteristic of a material flowing through the conduit in response to determining a frequency of a signal from the first pick-off sensor and a signal from the second pick-off sensor. 27. The apparatus (20) of claim 26, wherein one of the characteristics is a mass flow rate of material flowing through the conduit. 27. The apparatus (20) of claim 26, wherein one of the characteristics is density. 17. The system of claim 16, By performing (401) the first decimation from the first sample ratio to the intermediate sample ratio in response to receiving the signals from the first and second pick-off sensors, (20) configured to decimate a signal from the first pick-off sensor and a signal from the second pick-off sensor. 30. The apparatus of claim 29, Demodulate (403) the signal into an I component and a Q component; And By performing (404) a second decimation of the sample of the signal with a final sample ratio at the intermediate sample ratio in response to demodulation of the signal, And configured to decimate the signal. 16. The apparatus of claim 16, wherein the meter is a flow meter electronic element for a Coriolis flow meter.